Cytosine methylation is an epigenetic modification of DNA that plays a key role in cellular memory throughout the eukaryotes. It is important in several epigenetic gene regulatory systems including parental genomic imprinting, X-chromosome inactivation, and the silencing of transposons and other multiple copy DNAs. Methylation also plays a role in tumor cell biology, as tumors often show both genome wide demethylation and hypermethylation of specific genes. Despite its importance, little is known about how proper methylation patterns are established and maintained. The long-term goal of the proposed research is to use Arabidopsis thaliana as a model genetic system to identify and study genes involved in proper genomic methylation patterning. Arabidopsis is one of the best organisms for studies of this type, given the ability to do large-scale mutagenesis experiments, given the international Arabidopsis genome sequencing efforts, and given the collection of methylation mutants and cloned DNA methyltransferase genes which are already available. We have discovered a manipulable epigenetic system in which the methylation state of two key floral regulatory genes, SUPERMAN and AGAMOUS, can be altered by several mutations that affect overall genomic methylation. Surprisingly, we find that mutations that reduce overall genomic methylation, cause dense hypermethylation of SUPERMAN and AGAMOUS. Furthermore, we have discovered a new class of mutations that also cause dense overmethylation of the SUPERMAN locus, but that do not affect overall genomic methylation levels. The specific objectives of this proposal are 1) to map and study mutations which cause hypermethylation of SUPERMAN, with the goal of cloning the corresponding genes, 2) to use large scale mutagenesis to identify genes that are important in maintaining the dense patterns of methylation at SUPERMAN, 3) to study the role of SUPERMAN methylation during normal development, and 4) to determine cis acting sequences which cause targeting of SUPERMAN for methylation. Given the likelihood that genomic methylation mechanisms are conserved between plants and mammals, studies of plant gene hypermethylation may well shed light on similar processes occurring in animals, especially the superficially similar phenomenon of hypermethylation of genes in cancerous cells.